Despite a general convergence of findings implicating impairments of the frontostriatal regions and/or the brainstem catecholamine systems that innervate these networks in the pathophysiology of ADHD, the specific neural basis of the disorder remains elusive. Even less is known about the mechanisms by which medication treatments for ADHD exert their effects. The fact that stimulants are highly efficacious in reducing ADHD symptoms, coupled with their known effects on dopamine (DA) activity in the basal ganglia, provide strong evidence that DA mechanisms are highly relevant to ADHD. However, virtually all medications known to improve ADHD symptoms facilitate noradrenergic neurotransmission, and the efficacy of atomoxetine, a specific noradrenaline (NA) reuptake-inhibitor, demonstrates the importance of NA mechanisms in ADHD. Moreover, recent research in rodents indicating that both stimulants and atomoxetine directly influence DA and NA neurotransmission suggests that there are likely similarities as well as differences in their mechanisms of action. Better delineating how these treatments work is an important first step in matching treatments to patients, and would substantially enhance our understanding of the pathophysiology of ADHD. We are currently in an excellent position to use functional magnetic resonance imaging (fMRI) to better understand the neurobiological mechanisms of pharmacological treatments for ADHD. We were recently funded to study the comparative and differential effects of four to six weeks of methylphenidate and atomoxetine treatment in 160 children and adolescents with ADHD, using a cross-over design in which each subject receives both treatments (R01 MH070935-01). The current application proposes to conduct fMRI scans in 60 children 7-12 years (n=30 for each treatment) pre- and post-treatment, in the first medication arm of the parent cross-over study. The addition of fMRI scanning to this treatment study would provide a unique opportunity to evaluate whether proposed similarities and differences in actions between methylphenidate and atomoxetine derived from animal studies can be documented in children with ADHD, and to examine the degree to which clinical response to the two treatments is accounted for by shared or distinct mechanisms of action. We will also collect pilot data on neurobiological predictors of differential response. [unreadable] [unreadable] [unreadable]